MA1900 Midterm Exam Name: March 16, 2006 Instructions: Please answer each question to the best of your ability. There is one correct answer per question. If you have concerns about the answers that are supplied or think that there may be more than one answer, please explain why you believe this to be true. Good Luck!! and have a great Spring Break! 1. The drug manufacturer Merck recently stopped testing a promising new drug to treat depression. It turned out that in a randomized, double-blind trial a dummy pill did almost as well as the new drug. The fact that many people respond to a dummy treatment is called A) confounding. B) nonresponse. C) comparison. D) the placebo effect. 2. An experiment compares two brands of automobile tires. Each of a number of cars is equipped with one tire of each brand on a rear wheel (the order is randomized from car to car) and tread wear is measured periodically. This is called a A) simple random sample. B) stratified random sample. C) completely randomized design. D) matched pairs design. E) double-blind design. Use the following to answer questions 3-4: A newspaper article was headlined, "Marrying young can be beneficial." The article said: Marriage is a good thing when it comes to cutting back on drinking and drugs. A University of Michigan Institute for Social Research study of 33,000 young adults showed that young, unmarried adults usually increased their alcohol, marijuana and cocaine use when they left home, often to attend college. Those same people, however, decreased their drug and alcohol use when they got engaged or married. Couples who lived together but were not engaged or married showed no such drop in drug use. Page 1
3. The strongest reason for having reservations about the claim made in the first sentence of the quote is A) you don't agree with the claim. B) the sample size is too small. C) the study was limited to Michigan. D) there is no control group. E) it's an observational study. 4. The explanatory variable in this study is A) the level of alcohol, marijuana and cocaine use. B) marital status. C) year in college (e.g., freshman, sophomore, etc.). D) not given in the quote above. E) There is no explanatory variable because this is an observational study. Use the following to answer questions 5-9: A national sample survey interviewed 3,800 people age 18 and older nationwide by telephone. One question asked was whether they agreed with this statement: "Some people say we should have a third major political party in this country in addition to the Democrats and Republicans." 5. The population for this sample survey appears to be A) all adult residents of the U.S. B) all registered voters. C) the 3,800 people who were interviewed. D) people who think we should have a third political party. E) all U.S. citizens. 6. 53% of the people asked agreed that we should have a third party. The number 53% is a A) correlation. B) parameter. C) margin of error. D) confidence level. E) statistic. Page 2
7. The news article that reports the poll results says, "The margin of error is plus or minus two percentage points." This means that A) we can be sure that between 51% and 55% of all Americans think we should have a third party. B) we can be 95% confident that 53% of all Americans think we should have a third party. C) we can be 95% confident that between 51% and 55% of all Americans think we should have a third party. D) we can be 95% confident that between 51% and 55% of the 3,800 people interviewed think we should have a third party. E) we can be between 93% and 97% confident that 53% of all Americans think we should have a third party. 8. Among the poll respondents who were between 18 and 24 years old, 60% agreed that we should have a third political party. The margin of error that goes with this result is A) less than ± 2%, because there are fewer than 3,800 people 18 to 24 in the sample. B) greater than ± 2%, because there are fewer than 3,800 people 18 to 24 in the sample. C) equal to ± 2%, because this result comes from the same sample survey. D) less than ± 2%, because only some of the population is between 18 and 24, so it's easier to estimate facts about them. E) greater than ± 2%, because only some of the population is between 18 and 24, so it's harder to estimate facts about them. 9. A possible source of nonsampling error in this sample survey is A) some people chosen for the sample refused to answer questions. B) people without telephones could not be in the sample. C) some people never answered the phone in several calls. D) Both (a) and (c). E) All of (a), (b), and (c). Use the following to answer questions 10-11: "Congress passed a ban on the sale of assault weapons. Now there is a move to repeal that ban. Do you agree that the ban on sale of assault weapons should be repealed?" You ask that question to an SRS of 1000 adults in Texas (population 18 million people) and to a separate SRS of 1000 adults in Indiana (population 5.7 million people). You make 95% confidence statements about the percents of all adults in both states who agree. Page 3
10. Your margin of error for Indiana is A) the same as in Texas, because the two SRS's are the same size. B) larger than in Texas, because there are fewer people in Indiana. C) smaller than in Texas, because there are fewer people in Indiana. D) may be either smaller or larger than in Texas, because the sample result varies due to chance. E) smaller, because everything is bigger in Texas. 11. It would be cheaper to just announce the question on TV and ask people to call in with their opinions. Is this the best approach? A) This is a good way to get information about the opinions of all adults. B) This is a bad idea, because if only a few people call in your margin of error will be large. C) This is a bad idea, because allowing people to volunteer may result in large bias. D) This is a bad idea, because there is no control group. E) This is a bad idea, because the study isn't double-blind. Use the following to answer questions 12-17: Does taking large amounts of vitamins protect against cancer? To study this question, researchers enrolled 29,000 Finnish men, all smokers over the age of 50. Half of the men took vitamin supplements, and others took a dummy pill that has no active ingredient. The researchers followed all the men for eight years. At the end of the study, men in the vitamin group were no less likely to have cancer than men in the other group. This study cast doubt on the popular idea that taking lots of vitamins can reduce the risk of cancer. The study design looked like this: 12. The statistical name for this study design is A) simple random sample. B) stratified random sample. C) randomized comparative experiment. D) multistage sample. E) observational study. Page 4
13. The method used to form the groups should appear in the outline at the point marked (Question A). What is this method? A) Random allocation. B) Voluntary response. C) First come, first served. D) The double-blind method. E) Stratified sampling. 14. Treatment 2 was a dummy pill. Such a dummy treatment is called a A) double blind. B) categorical variable. C) nonsampling error. D) placebo. E) comparative. 15. The response variable should be named in the outline at (Question C). The response variable in this study is A) whether or not a subject took vitamins. B) 29,000 Finnish men. C) random allocation. D) a confidence statement. E) whether or not a subject developed cancer. 16. In order to avoid unconscious bias, neither the subjects not the doctors who examined them knew whether a particular subject was taking vitamins or dummy pills. This is called A) the placebo effect. B) the double-blind technique. C) the retrospective method. D) stratified sampling. E) internal validity. 17. A weakness of this study is that A) it isn't clear that the results apply to women. B) observational studies give only weak evidence for causation. C) the people who took vitamins may have had other good habits. D) the response is measured in a biased way. E) nonsampling errors may be large. Page 5
18. A company database contains the following information about each employee: age, date hired, sex (male or female), ethnic group (Asian, black, Hispanic, etc.), job category (clerical, management, technical, etc.), yearly salary. Which of the following lists of variables are all categorical? A) age, sex, ethnic group. B) sex, ethnic group, job category. C) ethnic group, job category, yearly salary. D) yearly salary, age. E) age, date hired. 19. A company used to give IQ tests to all job applicants. This is now illegal because IQ is not related to the performance of workers in all the company's jobs. That is, IQ as a measure of future performance on the job is A) biased. B) invalid. C) imprecise. D) unreliable. E) a sampling error. Use the following to answer questions 20-21: The Gallup poll interviews 1600 people. Of these, 18% say that they jog regularly. The news report adds that "The poll had a margin of error of plus or minus three percentage points." 20. You can safely conclude that A) 95% of all Gallup poll samples like this one give answers within ± 3% of the true population value. B) the percent of the population who jog is certain to be between 15% and 21%. C) 95% of the population jog between 15% and 21% of the time. D) we can be 3% confident that the sample result is true. E) if Gallup took many samples, 95% of them would find that exactly 18% of the people in the sample jog. 21. How many of the 1600 people in the Gallup Poll sample said that they jog regularly? A) 1312 B) 288 C) 89 D) Between 240 and 336, because of the margin of error. E) We can't say exactly, because we are only 95% confident. Page 6
22. Amy wants to spend a summer in France after she graduates. She is worried that this will be too expensive because the dollar buys fewer French francs now. When Amy entered college, a dollar was worth 6.5 francs. Now a dollar is worth only 5.25 francs. By about what percent has the value of the dollar in francs decreased? A) 10% B) 19% C) 24% D) 76% E) 81% 23. All statistical studies that collect data from people should follow some basic ethical standards. Some of these ethical standards are: A) Always use random selection to choose among people. B) Keep data on individuals confidential. C) Get informed consent before collecting data. D) Both (B) and (C). E) All of (A), (B), and (C). 24. You want to ask a sample of professors at your school how they feel about the tenure system for faculty. You realize that opinions may differ depending on the rank of the professor, especially because assistant professors do not have tenure. So you take separate SRS's of assistant, associate, and full professors and combine them to form your sample. You used a A) simple random sample. B) stratified random sample. C) multistage sample. D) completely randomized experiment. E) randomized block experiment. Use the following to answer questions 25-26: Were the extinctions that occurred in the last ice age more frequent among species of animals with large body sizes? A researcher gathers data on the average body mass (in kilograms) of all species known to have existed at that time. 25. These measurements are values of A) a categorical variable. B) a quantitative variable. C) an invalid variable. D) a margin of error. Page 7
26. What are the explanatory and response variables? A) There is no explanatory-response distinction in this situation. B) Explanatory: body mass of a species. Response: whether the species went extinct. C) Explanatory: the ice age. Response: whether a species went extinct. D) Explanatory: whether a species went extinct. Response: the body mass of the species. E) Explanatory: the ice age. Response: the body mass of a species. Use the following to answer questions 27-29: You want to know the opinions of American school teachers about establishing a national test for high school graduation. You obtain a list of the members of the National Education Association (the largest teachers' union) and mail a questionnaire to 2500 teachers chosen at random from this list. In all 1347 teachers return the questionnaire. 27. In this situation, the population is A) the 1347 teachers who mail back the questionnaire. B) the 2500 teachers to whom you mailed the questionnaire. C) all members of the National Education Association. D) all American school teachers. E) all American school students. 28. The sampling frame is A) the 1347 teachers who mail back the questionnaire. B) the 2500 teachers to whom you mailed the questionnaire. C) all members of the National Education Association. D) all American school teachers. E) all American school students. 29. The sample is A) the 1347 teachers who mail back the questionnaire. B) the 2500 teachers to whom you mailed the questionnaire. C) all members of the National Education Association. D) all American school teachers. E) all American school students. Page 8
30. An example of a nonsampling error that can reduce the accuracy of a sample survey is: A) Using voluntary response to choose the sample. B) Using the telephone directory as the sampling frame. C) Interviewing people at shopping malls to obtain a sample. D) Variation due to chance in choosing a sample at random. E) Many members of the sample cannot be contacted. 31. Any sample survey should follow these two basic principles of data ethics: A) Anonymity and random sampling. B) Comparison and randomization. C) Double-blind and placebo. D) Bias and variability. E) Informed consent and confidentiality. 32. In a table of random digits it is true that A) every pair of digits has chance 1/100 to be any of the 100 possible pairs 00, 01,..., 99. B) if a pair of digits is 00, the next pair cannot also be 00. C) every row has exactly the same number of 0's and 1's. D) (A) and (B) but not (C). E) All of (A), (B), and (C). 33. A radio talk show invites listeners to call a telephone number to vote "Yes" or "No" on whether they support a bond issue for a new school. About 1500 people call in. Over 80% say "No." As an estimate of community opinion, this result is A) accurate to within ± 3% with 95% confidence. B) not trustworthy because of nonsampling errors. C) not precise because the sample size 1500 is too small. D) unethical due to lack of informed consent. E) badly biased due to voluntary response. 34. Corn variety #1 yielded 140 bushels per acre last year at a research farm. This year, corn variety #2, planted in the same location, yielded only 110 bushels per acre. Unfortunately, we don't know whether the difference is due to the superiority of variety #1 or to the effect of this year's drought. This is an example of A) bias due to voluntary response. B) random sampling error. C) confounding. D) the placebo effect. E) nonsampling error. Page 9
35. The Environmental Protection Agency sends four identical samples of polluted water to the Acme Chemical Analysis Laboratory. The lab measures the amount of dioxin in each sample. The lab reports dioxin concentrations of 141, 323, 74, and 923 parts per billion. You can see that dioxin measurements made by Acme are A) not reliable. B) not biased. C) invalid. D) confounded. E) categorical. 36. The most important advantage of experiments over observational studies is that A) experiments are usually easier to carry out. B) experiments can give better evidence of causation. C) confounding cannot happen in experiments. D) an observational study cannot have a response variable. E) observational studies cannot use random samples. 37. The net asset value of a mutual fund has increased from $27 on December 31 to $33 now. The percent increase in value is about A) 22%. B) 18%. C) 1.2%. D) 122%. E) 82% 38. It is hard to measure "intelligence." Let's do it the easy way: measure height in inches, and call the result "intelligence." Not only is this method easy, it gives the same number every time we repeat the measurement on the same person. Measuring intelligence this way is A) not reliable and not valid. B) highly reliable but not valid. C) valid, but not reliable. D) both valid and highly reliable. Page 10
39. Any statistical study that uses human subjects requires informed consent. This means that A) a review board must agree that the study is worthwhile and will not harm the subjects. B) the authors of the study must agree to inform the public of the study results. C) the authors of the study must inform a review board about it and obtain the board's consent to go ahead. D) the nature of the study must be explained in advance to the subjects and they must voluntarily agree to take part. 40. In a clinical trial to study the effectiveness of a vaccine in preventing a new disease, the best design is to A) give the vaccine to the high-risk subjects and a placebo to the rest. B) give the vaccine to those who want it, and use the others as a control group for comparison. C) give the vaccine to all who sign a consent form. D) choose a control group and a treatment group by random allocation from among those who consent to take part in the study. Page 11